1 Lecture Notes PH 411/511 ECE 598 A. La Rosa INTRODUCTION TO QUANTUM MECHANICS ________________________________________________________________________ CHAPTER-7 THE CONCEPT of AMPLITUDE PROBABILITY 7.1 Quantum mechanics description in terms of base-states 7.2 Probability amplitude 7.3 General Guiding principles to assign amplitude probabilities 7.4 Interference between Amplitude Probabilities 7.4.A Two-slit experiment: Watching electrons’ trajectories Adding amplitude-probabilities for events producing the same final state. Adding probabilities for vents producing different final states. 7.4.B Neutron Scattering from a crystal by particles with or without spin 7.4.B.a Angular Momentum and Magnetic Dipole Orbital angular momentum The Stern-Gerlach experiment: Measurement of the magnetic dipole . Bohr’s quantization of the angular momentum. Electron Spin, spin S . Spin gyromagnetic ratio Nuclear spin spin I 7.4.B.b Scattering form a crystal of atoms with spin zero 7.4.B.c Scattering form a crystal of atoms with spin ½. References R. Feynman, “The Feynman Lectures on Physics,” Volume III, Chapter 3. R. Eisberg and R. Resnick, “Quantum Physics,” 2nd Edition, Wiley, 1985. Sections 8-1 to 8-3. B. H. Bransdem& C.J. Joachin, “Quantum Mechanics” 2 nd Ed. Prentice Hall, 2000. Section 1.5.
30
Embed
CHAPTER-7 THE CONCEPT of AMPLITUDE PROBABILITY · PDF fileTHE CONCEPT of AMPLITUDE PROBABILITY ... 7.4.B.b Scattering form a crystal of atoms with spin zero ... The Feynman Lectures
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Lecture Notes PH 411/511 ECE 598 A. La Rosa
INTRODUCTION TO QUANTUM MECHANICS ________________________________________________________________________
CHAPTER-7 THE CONCEPT of AMPLITUDE PROBABILITY
7.1 Quantum mechanics description in terms of base-states
7.2 Probability amplitude
7.3 General Guiding principles to assign amplitude probabilities
7.4 Interference between Amplitude Probabilities
7.4.A Two-slit experiment: Watching electrons’ trajectories Adding amplitude-probabilities for events producing the same final state. Adding probabilities for vents producing different final states.
7.4.B Neutron Scattering from a crystal by particles with or without spin
7.4.B.a Angular Momentum and Magnetic Dipole Orbital angular momentum
The Stern-Gerlach experiment: Measurement of the magnetic
dipole
.
Bohr’s quantization of the angular momentum.
Electron Spin, spinS
. Spin gyromagnetic ratio
Nuclear spin spinI
7.4.B.b Scattering form a crystal of atoms with spin zero
7.4.B.c Scattering form a crystal of atoms with spin ½.
References
R. Feynman, “The Feynman Lectures on Physics,” Volume III, Chapter 3.
R. Eisberg and R. Resnick, “Quantum Physics,” 2nd Edition, Wiley, 1985. Sections 8-1 to 8-3.
B. H. Bransdem& C.J. Joachin, “Quantum Mechanics” 2nd Ed. Prentice Hall, 2000. Section 1.5.
2
CHAPTER-6 THE AMPLITUDE PROBABILITY1
In Chapter 5 we emphasized the crucial aspect played by the measurement process in Quantum Mechanics, leading to the enunciation of the Heisenberg uncertainty principle. The determinism (characteristic of classical physics) was replaced by the probabilistic character (characteristic of quantum physics) to predict the behavior of a system of particles.
One striking new feature in quantum mechanics refers to the impossibility in obtaining a perfect knowledge of all the physical variables that relate to the motion of an atomic object (electrons, in the case of the experiment to be described below). There is always an inherent uncertainty in the value of the variables (position, velocity, …) describing the dynamic of the object. Pictorially, there is an absence of trajectories.
At the beginning, we may get the impression that such lack of determinist description of the electron’s motion has the appearance of a mystery. This mysterious (no deterministic) description of the atomic phenomenon may constitute the only ‘mystery’ in quantum mechanics. Thus, it is worth to consider having an early exposure to it
through a “simple” experiment of electrons passing through two slits.
A warning must be stated, however; paraphrasing Feynman,
we will not be able to explain how it works, we will tell you how it works.1
That is to say, we will provide (this will actually be done in Chapter 8) a
mathematical procedure (based on the concept of amplitude probabilities)
that when applied to a given problem (an electron passing through a couple of slits, for
example) it predicts an outcome that happens to coincide with the
experimental results (this will be done in terms of “amplitude probability”
concept). But we may still leave you wondering “what does exactly the electron do”?
The wavefunction itself fully characterizing a given system (although, to be more
precise, recall that represents the associated ensemble of identically prepared
systems) does not have a direct interpretation in classical terms. It is 2
ψ what is
interpreted as probability (the latter being a concept we can be understood in classical
terms). To gain some familiarity with itself and to learn how to mathematically manipulate it in order to make predictions on the behavior of a system, in this chapter
we will give emphasis to the description of itself and its spectral components (relative to a given base states.)
3
7.1 Quantum Mechanics Description in terms of Base-states
The concept of spectral components was introduced in Chapter 4:
Given a pre-determined basis-set of functions
...},2,1;{ nn ,
which satisfy
mnmn ,
then, any state of motion of the system can be described by a proper linear combination of the form,
n
n
n A ψ
State Base State .
(1)
Notice the generality of the notation, where the detailed dependence of on the coordinates x and t is put aside. Generality is gained because it allows the study of almost any quantum mechanical system, including the ones that do not admit a description in terms of spatial coordinates such as x. An example of the latter systems are the ones that involve spin (a property to be described at the end of this chapter.)
If we want to obtain An,
n ψ gives An
(n ,
n │
')'( )'( *
n dxxx
4
Just different notations for the same quantity
Let’s recall also that, given two arbitrary complex functions and ψ , the scalar
product between and ψ is defined as follows,
dxxx )(ψ)(ψ *definition of scalar product.
More about the scalar product, bra and ket notation Before engaging in the description of amplitude probabilities, lets review a subtle manipulation of the inner product definition, which it will widely used in this and future chapters.)
Given n
n
n A ψ and m
m
m B ,
how to calculate ? (2)
We do have but we need
so we can then calculate
Can we just affirm that:
m
m
m B implies m
m
m B ? (3)
If that statement were correct, we would obtain
=
m
mm
B
n
nn
A = m
nmnm AB n
However, this last expression is incorrect.
To find out the correct value of let’s use the original notation that uses
parenthesis (as defined in Chapter 4).
Taking n
n
n A ψ and m
m
m B , the definition of the scalar product
))(ψ)(ψ, * dxxxχ leads us to the following,
5
dxAxBxABχ n
n
nm
m
mn
n
nm
m
m )()( ,ψ,
*
=
dxAxBx n
n
nm
m
m )( )( **
dxAxBx nnmm
m n
)()(**
dxxxAB nm
m n
nm )( )(**
nm
m n
nm AB ,*
Thus,
n
n
n A ψ and m
m
m B implies
nm
m n
nm ABχ ,ψ,*
(4)
Translating this result (4) into bra-ket notation,
If m
m
m B and n
n
n A ψ then,
n m
nmnm AB ,*
alternatively
n
n
n
m
mm AB *
(5)
This suggests the following rule:
Ket Bra
6
m
m
m B
m
mmB *
(6)
where *
mB stands for the complex conjugate of mB .
Exercise: Bra Ket
χχ
Ψa
Ψa Ψa
that Show
Φ
ΦΦ
*b
b b
that Show
Exercise: Find the corresponding ket of the following bra Φ8i -
Solution Long justification
Φ8i - = )( * Φ -8i = )( Φ 8i
)( Φ 8i
Φ 8i *)(
Φ 8i )(- = Φ )(-8i
Short justification Φ8i -
Φ )(-8i
= Φ 8i )(-
7.2 Definition of Amplitude Probability
In the expansion of ψ in terms of the base states { n , n= 1, 2, …}, given by
n
n
n A ψ , the complex number coefficients An (to be referred as amplitude
probabilities) are obtained by means of the scalar product,
nn A Amplitude probability
Interpretation:
7
In this formalism, the probability amplitude nn A is a complex number whose
magnitude squared 2*2
nnnn AAA gives the probability that a system
initially in the state ψ will be found in the base state n after a measurement.
Generalization:
In a more general context,
Amplitude Probability
Complex number whose magnitude squared 2
gives the probability that a system initially
in the state will be found in the state after
the measurement.
(7)
7.3 General guiding principles to assign amplitude probabilities
To familiarize with the proper mathematical manipulation of amplitude probabilities, we’ll follow Feynman’s general guiding principles. For illustrative purposes, these principles are presented here along the description of the experiment described in Chapter 6 where particles (photons or electrons) pass through a couple of slits. It was mentioned there that we could not determined which aperture a particle uses before reaching the detection screen; when attempting to watch their trajectory, the mere observation affected dramatically the result. We would like to formalize the description of that experiment in terms of the amplitude probability tool.
Principle-1 How to assign amplitude probabilities
The amplitude probability that a particle leaving the source s will arrive at the position x on the screen can be represented by a complex number as follows,
sx leavesparticleatarrivesParticle (8)
To simplify the notation of this complex number in the particular case that we are considering, instead of (8) let’s use,
sx (9)
8
1
2 Filament (Electron source)
Wall Observation screen
Detector
s x
Fig. 7.1 Schematic set up of the two-slit experiment.
For a more general case, this notation will imply
ψstatestate startingfinal (10)
Amplitude probability that a system in the
starting state “” ends up in the final state “”.
Principle-2 When to add amplitude probabilities When a particle can reach a given final state by two (or more) possible routes, the total amplitude for the process is the sum of the amplitudes associated to the two (or more) routes separately.
In other words, we add amplitudes when the corresponding possible routes have the same initial state and the same final state.
In our example, since the event consisting of a particle leaving “ s ” and arriving at “ x ” can be realized, via the aperture 1, or
via the aperture 2,
each path having the same intial state and the same final state. Therefore the amplitude
sx will be given by,
9
2via1via sxsxsx (11)
Both paths have the same initial and final states
Principle-3 When to multiply amplitude probabilities
The amplitude for a given route can be written as the product of i) the amplitude to go part of the way, times ii) the amplitude to go the rest of the way.
In our particular case, the amplitude to go from s to x via the aperture-1 will be written as,
sxsx 1 11via
(12)
Based on (12), expression (11) can be re-written as,
s x s xsxopen apertures both
2211 (13)
We may complain that we do not know exactly what are the values of the terms we
are dealing with (like, for example, what is the numerical value of sx 11 .) But be
patient at this stage. We are not ready yet to calculate numerical values; rather we are focused in learning how to operate with those numbers. [We do not know what the lunar module is made of; but we are learning how to operate it in our quest to land on the Moon.]
7.4 Interference between Amplitude Probabilities
In this section we describe some specific examples that illustrate the simplicity and power of the amplitude probability formalism to correctly predict and understand quantum phenomena (in particular those that are not possible to explain with classical mechanics tools.) The key aspect to learn in this section is the occurrence of interference between amplitude probabilities terms associated to different events that produce the same final state. The first example is related to the two-slit experiment encountered in Chapter 6, which was a bit strange to understand with classical arguments. Being familiar with that experiment will help us to understand how to apply the concept of amplitude probabilities. The second example deals with the scattering of
10
neutrons from a crystal, and emphasizes the role played by the spin character of the particle.
Chapter 6 described how the act of using photons to watch the incident electrons, aiming to distinguish which of the two apertures a given electron passed through, affected substantially the pattern of intensity on the observing screen. In other words, the pattern of intensity on the screen is dramatically affected by the act of observing the electrons with a source of light. That example was used to illustrate the Heisenberg’s uncertainty principle, which reflects the inherent limitations an observer must face when trying to measure the variables characterizing the motion of a particle. The act of measurement perturbs the system under analysis thus preventing the knowledge on the values of the physical parameter before the measurement.
Armed with the tool of amplitude probability (and its associated rules), let’s describe the two slit experiment under such formalism and check what predictions it makes under different circumstances.
P1
X
P2
P12
X
1
2 Filament (Electron source)
Wall Observation
screen
s
D2
D1
Photon counter
Photon counter
x L
Fig. 7.2 Two-slit experiment setup where electrons are being watched by photons (the latter monitored by two photon detectors.)
“Amplitude probability” description of the two-slit experiment
A light source “ ” is used in an attempt to watch what specific aperture an electron Lchooses to pass through.
Simplification: We will assume that when an electron passes through a given aperture it
will scatter a photon towards either the photon-detector D1 or the photon detector D2.
(That is, we assume a scattered photon does not go undetected.)
There are different possible results. Let’s analyze them in more detail.
11
Monitoring the experiment with the photon counter D1
Event: Electron starts at “s”, ends up at the position “x” on the screen, and a photon is detected at counter D1
What is the amplitude probability for this event to happen?
Path-1. There is an amplitude probability that an electron goes from the source s
to the aperture-1: 1s
While the electron passes through the slit-1, there is an amplitude probability that it
scatters a photon into the photon counter D1; let’s call this amplitude a
There is an amplitude probability that the electron goes from the slit-1 to the
position x on the screen: x1
Accordingly the amplitude probability that an electron goes from s to x via the aperture-1, AND scatters a photon into the photon counter D1 is
given by: x1 a 1s
Path-2. An electron passing through the slit-2 could produce the same result. That is,
There is an amplitude probability that an electron goes from the source s to the
aperture-2: 2s
While the electron passes through the slit-2, there is an amplitude probability that it
scatters a photon into the photon counter D1; let’s call this amplitude b.
[It is possible that the magnitude of b could be very small compared to the
magnitude of a, since the photon detector D1 is farther away from aperture-2.]
There is an amplitude probability that the electron goes from the slit-2 to the
position x on the screen: x2 .
Accordingly: the amplitude probability that an electron goes from s to x via the aperture-2, AND scatters a photon into the photon counter D1 is
given by: x2 b 2s
Interference between events that produce the same final state:
Notice in the two cases mentioned above that the net result is an electron going
from s to x , AND a photon from the source L scattered into the photon counter D1. That is, the final state produced by these two events can not be distinguished.
12
Accordingly, we have to add up these two amplitudes to describe the following event:
Lfromphoton
sfromelectron
Dinphoton
xatelectron
1
= 1x a s1 + 2x b s2 (14)
These two events have the same initial state and follow a path that (respectively) leads to the same final state. Accordingly, we add up their corresponding amplitude probabilities.
Monitoring the experiment with the photon counter D2
Event: Electron starts at “s”, ends up at the position “x” on the screen, and a photon is detected at counter D2
What is the amplitude probability for this event to happen?
By symmetry, the amplitude that an electron scatters a photon into the photon
counter D2 while passing through the aperture-2 should be equal to a .
Also, the amplitude that an electron scatters a photon into the photon counter D2
while passing through the aperture-1 should be equal to b . Accordingly,
Lfromphoton
sfromelectron
Dinphoton
xatelectron
2
= 2x a s2 + 1x b s1 (15)
These two events have the same initial state and follow a path that (respectively) leads to the same final state. Accordingly, we add up their corresponding amplitude probabilities.
Expressions (14) and (15) should be compared with result (13) above
sxsxsx 2211 (obtained for the case where no light source was
used). Notice that,
the effect of using light introduces “weighting” complex factors a and
b into the participating amplitudes sx 11 and sx 22 .
13
To understand the effect of the light source, let’s calculate the probability for different situations.
Example: When to add “probabilities” and not “amplitude-probabilities” Event: Electron starts at “s”, ends up at the position “x” on the
screen (regardless whether or not the scattered photon ended up at D1 or D2)
(16)
What is the amplitude probability for this event to happen?
The event described in (16) appears to be composed of the two events treated in the section above. Accordingly,
should we add the amplitude probabilities given in (14) and (15)?
Paraphrasing Feynman, the answer is NO!
You must never add amplitudes corresponding to events that have distinct final states.
Indeed, compare (14) and (15) and notice that they refer to distinct final states. In one case the photon went to D1; in the other case the photon went to D2.
Lfromphoton
sfromelectron
Dinphoton
xatelectron
1
Lfromphoton
sfromelectron
Dinphoton
xatelectron
2
These two final states are different
We may say,
“Why not just ignore the information available from the photon counters. That way we have a situation similar to the one when we did not have the light source nor the photon counter.”
But, looking or not to the photon counters “is your business … Nature does not know what you are looking at, and behaves the way she is going to
behave whether or not you bother to take a look to the photon counters. …”
“Once a photon is accepted by one of the photon counters, we can always determine which alternative occurred if we want, without causing any further disturbance to the system or to the experiment.”
14
(This latter expression from Feynman is key!!!) That is, we can find out which of the two cases occurred without disturbing the experiment.
Accordingly, in the two-slit experiment with a light source present in the set up we shouldn’t be asking about the “amplitude-probability” that an electron arrives at x regardless of which photon counter the photon went;
instead we should ask abut the “probability” for such an event to occur.
Calculation of “probability” instead of “amplitude probability” applies in the case indicated in (16) because it involves a collection (in this case two) of different final states.
(17)
Accordingly, the probability that an electron leaving from s arrives at x (while being watched by photons) is given by :
22
22
22112211
)(
2
1
saxsbxsbxsax
sx;
Lfromphoton
sfromelectron
Dinphoton
xatelectron
Lfromphoton
sfromelectron
Dinphoton
xatelectron
presentissourcelighawhenP
(18)
For comparison purposes, let’s re-write expression (13), which gives the probability that an electron leaving from s arrives at x (without being perturbed by photons), in the following way,
2 2
2211
)(
sxsxsxpresentsourcelightno
presentsourcelightnosx;P
(19)Thus,
presentsourcelightpresentsourcelightno
sx;sx; PP )( )( 2
15
Analyzing particular cases
Case: Aperture-2 is closed
222
22
11)(
1111);(
sxba
sbxsaxsxP
Notice, no interference terms.
P1
X
1
2 Filament (Electron source)
Wall Observation
s
D2
D1
Photon counter
Photon counter
x
Fig. 7.3 Only one aperture open. As the electron passes though the aperture 1, a photon
can end up at D1 or D2. The latter contributes to the probability with a factor )(22
ba
Case: Aperture-1 is closed
222
22
22)(
2222);(
sxba
saxsbxsxP
Notice, no interference terms.
X
P2
1
2 Filament (Electron source)
Wall Observation
s
D2
D1
Photon counter
Photon counter
x
Fig. 7.4 Only one aperture open. As the electron passes though the aperture 2, a photon
can end up at D1 or D2. The latter contributes to the probability with a factor )(22
ba .
16
Case: Photon of very long wavelength (i.e. p=h/ very small) as to minimize the perturbation of the electron’s motion.
For photons of long wavelength, the lateral resolution becomes so poor that a photon arriving at D1, for example, could equally come from either an electron passing through aperture 1 or aperture-2. Hence, we will not know through which aperture the electron passed through.
Alternatively we could say that the photons are of so low energy that they could easily be scattered in any direction by an electron; that is, a photon could end up with equal probability in the photon detector-1 or the photon detector-2. This leads us to consider
that the coefficients a and b should have approximately equal.
If we take ba (that is, amplitude probability that a photon is scattered into the photon counter D1 is the same as the amplitude probability that a photon is scattered into the photon counter D2), expression (18) becomes,
22
22
22
22112
2211
2211);(
sxsxa
sxsxa
sxsxasxP
Interference term
(20)
P12
X
1
2 Filament (Electron source)
Wall Observation P12 ≠ P1 + P2
screen
s
D2
D1
Photon counter
Photon counter
x
Fig. 7.5 Interaction with photons of very large wavelength. As the electron passes though the wall, a photon of very large wavelength interact with it. The wavelength is so large that when, for example, a photon arrives at D1 it could have come from aperture 1 or 2. That is, we will not know through which aperture the electron passed through. Notice, the consequence is that an interference pattern is formed on the screen.
17
Thus, when we do not know which aperture an electron passes through we get a wavy pattern of intensity on the screen.
Case: Using photons of small wavelength, as to improve the lateral resolution and thus identify which aperture an electron passed through.
One can argue that the energy of a photon is so high that a photon will be scattered more towards the forward direction; thus it is more likely that it will end up in the closer photon counter than in the farther away; hence we will know
which aperture the electron passed through. In such a case, we expect ba .
The phenomena will be then influenced mainly by a and minimally by b. Taking b=0 in (18) we obtain,
222
22
2211
2211);(
sxsxa
saxsaxsxP
(21)
X
P12 1
2 Filament (Electron source)
Wall Observation P12 = P1 + P2 screen
s x
D2
D1
Photon counter
Photon counter
Fig. 7.6 Interaction with photons of very small wavelength. This allows us to know which aperture the electron passes through, but at the cost that the interference pattern is destroyed.
We do know which aperture an electron passed through, and we get a pattern of intensity on the screen that is simply the addition of the probabilities P1 and P2 (no interference occurs.)
18
7.4.B Scattering from a crystal by particles with or without spin
7.4.B.a Angular Momentum and Magnetic Dipole
Orbital angular momentum
In a classical hydrogenic atom model:
An electron occupies a circular orbit, rotating with an orbital
angular momentum v
xrmL e
Since a moving charge gives rise to an electric current I, an
electron moving in a close orbit of radius r forms an equivalent
magnetic dipole
of magnitude ) ( 2)( rI
I
- e r
L
v
v
xrL em
Fig. 7.7 Orbital motion of an electron with associated angular momentum and magnetic dipole.
Let’s quantify these two quantities.
Magnetic dipole: 2 ))(( rareacurrent I
2 r
periodorbit
e
Orbital angular momentum: vrL e
m
odorbit peri
thorbit leng rvr
eemm
odorbit peri
rrvr
2
ee
mm
19
odorbit peri
r 2
2
em
From these expressions the following relationship is obtained,
L
em
e
2 (22)
7.4.B.b The Stern-Gerlach experiment: Measurement of the
magnetic dipole
.
Interaction with a Magnetic Field
A magnetic dipole
inside a magnetic field B
experiences a
force, whose components are given by,
, x
BF
x
,
y
BF
y
z
BF
z
(23)
In 1921, Stern suggested that magnetic moments
of atoms
could be measured by detecting the deflection of an atomic beam by
an inhomogeneous magnetic field B
.
The experiment was made using atoms of silver
A sample of silver metal is vaporized in an oven.
A fraction of atoms emerging from a small hole is collimated by the apparatus slits, which enter the magnetic field region.
The poles are designed in order to provide a more pronounced variation of the magnetic field in the z-direction.
(Notice in the schematic diagram above, that the change of B
along the x and y directions is much less, if not null, than in the
z-direction). Thus the net force acting on the dipole is given by,
z
B
z
BFz
zz
(24)
20
The atoms in the incident beam are deflected according to the value of their magnetic moment, and detected upon on a collecting plate.
Oven Oven
Magnet
Observed pattern
x
z
y
B
Fig. 7.8 Left: Schematic apparatus to measure the
magnetic moment of atoms. Right: Spatial distribution of the detected atoms.
Classical prediction
Since the orientation of the silver atoms in the incident beam is
random, the corresponding values of z should be spread over a
continuum range
–M < z < M, (25)
where M is the magnitude of
.
Consequently, the spots on the collecting plate should distribute over a continuum filled spot, centered on the non-deflection
direction.
However, this is not what is observed, as seen in the right side schematic of Fig. 7.8.
Bohr’s quantization of the angular momentum
A quantum interpretation could be explored according to the Bohr’s model of a hydrogenic atom. Bohr had postulated that the orbital
angular momentum L of the electron can take only discrete values
given by,
2
hnnL (26)
with n being the quantum number, ... 3 2 1 ,, ,n
and h is the Planck’s constant.
21
Since the angular momentum is, according to Bohr, given in units of
h, it is usual to express the magnetic dipole, given in expression (22)
above, in terms of the Bohr’s magneton B
Lμ
2 em
eorbital (27)
B = 10 9.27 Tesla
Joule24-x Bohr’s magneton
Accordingly, a discrete variation of L
would
imply a discrete variation of orbitalμ
.
Quantization of zL
If we further postulate that the component zL is also quantized
(“space quantization”), then zL would change within the range,
maxmax LLL z -
More specifically,
for a given orbit of angular momentum lL ,
the possible values of zL would be
mL z ,
with m taking values within the range,
llll , - , ... , , 11
(28)
Thus, for a given l the multiplicity for the possible
values of zL is 12 l
For l =0 the multiplicity is 12 l =1
l =1 12 l =3
l =2 12 l =5
However the results from the Stern and Gerlach experiment does not fit this scheme; the experimental result displayed in Fig 7.8 above
indicates a multiplicity equal to 2, which requires 2/1l .
Electron Spin, spinS
In 1925, S. Gousmit and G. E. Uhlenbeck suggested that, in addition
to the magnetic moment produced by the orbital motion, orbitalμ
,
22
Electrons could also posses an intrinsic magnetic
moment spinμ
, where its components in a given
direction can take only two values ± spin.
(Hence the split of the beam in two, after passing through the magnetic field, as shown in Fig 7.8.)
Associated with spin
one also postulates an
intrinsic angular momentum, or spin of the
electron, spinS
.
IF, in analogy to the orbital angular momentum l, we introduce a spin
quantum number s so that the multiplicity of spinz )(S along the z
direction be equal to 2s+1, THEN for an electron we must have s=1/2
The postulates mentioned above are summarized in the following expression (by analogy with expression (27)),
spin
spin
em
egs
S
2 (29)
with zS = 2
1 or
2
1 (for the case of one electron)
[If we had used an expression similar to (27) by simply replacing
orbital
and orbitalL
with spin
and spinS
respectively, the resultant
relationship
spin
spinm
e S
2 would predict an intrinsic magnetic
moment too small by a factor of 2, compared to what is measured
experimentally. Hence the need to ad-hoc introduce the factor sg .
The origin of sg is in fact a relativistic quantum effect.]
In the expression (29) above,
2.00232 sg (30)
and it is called the “spin gyromagnetic ratio.”
23
Using expressions (27) and (29), the total magnetic moment for an electron in an atom is give by,
spin
e
totalsgorbital
m
e SL
2
(31)
magneton sBohr' 10 9.27 Tesla
Joule24-
em
eB x
2
Nuclear spin spinI
Similar to expression (31), the nuclear magnetic moment nuclear
and
the nuclear spin spinI
of a given nucleon are related by,2
spin
nucleon
nuclear
m
eg
I
2 (32)
N nuclear magneton
eV/T .05084x10 -275
N
For a free proton: 2/ protonI , 5.58569 protong , Nproton 2.79284
For a free neutron: 2/ neutronnI , 82608.3 neutrong ,
Nneutron -1.9130418
Even the uncharged neutron has a large magnetic moment. This suggests that there is internal structure involving the motion of charged particles.
7.4.B.b Scattering from a crystal of atoms of spin zero
Consider,
a) A crystal: a periodic array of atoms with nuclei at their centers,
The atoms are labeled by the index i =1, 2, 3, … N.
b) A beam of neutrons incident on the crystal.
24
The objective is to calculate the probability that a neutron gets into the counter at a given angular location, as shown in the figure below.
Neutron counter
Crystal
C
S Neutron
source S
C
1 2 j N
Fig. 7.9 Schematic experimental arrangement for the diffraction of neutrons by a crystal.
Let’s assume the incident neutrons have a relatively low energy so that they do not cause any major disturbance or modification on the crystal (i.e. they do not leave track of their interaction.)
Scattering via atom i : The amplitude probability that a neutron arrives at C scattered by
the atom i in the crystal can be represented by,
< neutron at C neutron from S >via i = SiiC
S
C
1 2 i N Detector
located far
away
When a neutron arrives at the counter, it could have come scattered from any atom in the crystal, and there is no way to know which atom produced the scattering. We have therefore different routes with the
same initial and final state, which will therefore interfere as to give,
N
i
SiiCom Sneutron frCatneutron (33)
Since we are adding amplitudes of scattering associated to atoms at different space locations, the amplitudes will have different phases and will produce an interference pattern similar to the diffraction grating studied in chapter 6.
25
S
Source of neutrons
S
C
1 2 3 N
Crystal
Counts Diffraction by a crystal of
atoms of spin=0
Fig. 7.10 The neutron intensity as a function of the angle results from constructive and destructive interference of the different amplitudes associated to the different possible paths that join the same initial state (S) and the same final state (C).
The intensity pattern shown in Fig. 7.10 above, however, is different when using other crystals made out of atoms that do have spin. The spin character of the atoms will be reflected in the different intensity pattern response as explain below.
7.4.B.c Scattering form a crystal of atoms with spin 1/2
So far, we have not considered the fact that,
a neutron has spin ½ (34)
(as discussed in Section 7.4.B.a above.)
This gives two spatial directions to consider for the nuclear dipole moment:
up or down
(along, for example, the direction of the axis
perpendicular to the page in figure 7.9.)
CASE-1. If the nuclei of the crystal had no spin, the spin of the incident neutrons would have no effect in the experiment. Thus an intensity pattern similar to the one shown in Fig 7.10 would be obtained.
Counts Crystal of atoms with no spin
26
CASE-2. But, if the nuclei in the crystal had, for example, spin ½ a different intensity pattern is obtained due to the different possible interchange of spin between the nuclei and the incident neutrons.
1 2 3 N
Crystal
Counts How would the intensity pattern be modified?
?
Let’s consider CASE-2 in more detail.
In this discussion we will apply a rule similar to the conservation of angular momentum:
2A. If the neutron and the atomic nucleus have the same spin, then the scattering will produce no change in the spin of either particle. Before
After
S
C
1 2 3 N
Crystal
1 2 3 N
Crystal
This case would be similar then to the one discussed in section 7.4.B.b above (results displayed in Fig. 7.10.)
Counts spin
2B If the neutron and the atomic nucleus have opposite spin, then scattering can occur by two processes:
The spins of the neutron and atom remain the same
27
Before
After
S
C
the spin directions are exchanged
Before
After
S
C
We’ll consider these two possible cases separately.
Incident neutron and scatterer nuclei having opposite spin
Let’s define, a = amplitude probability that there is no
interchange of spin due to the scattering
b = amplitude probability that there is exchange of spin
(35)
To start gaining some understanding of the problem, we will assume that all the nuclei are set up with spin in one direction, pointing down for example. When the incident neutron has spin up, two different scenarios can occur:
CASE 2B.1: No interchange of spin
28
Initial state
Final state
S
C
1 2 3 N
1 2 3 N
Fig. 7.11
Amplitude probability that the neutron is scattered toward C via the
nucleus-i:
idnucleialldnucleiall
s via nucleuupup SC own own ;; = iC Sia (35)
Amplitude probability that the neutron is scattered toward C via the
nucleus-k:
kdnucleialldnucleiall
s via nucleuupup SC own ;own ; = kC Ska (36)
Notice, the two path-alternatives alluded in (35) and (36) leave the crystal intact; also, both paths, k and i, have the same initial and final state. In fact, the various possible path-alternatives (scattering from any of the N nuclei) produce the same initial and final state.
Accordingly:
Amplitude probability that the neutron is scattered toward C,
own own ;; dnucleialldnucleiall upup SC = Sa N
1
iii
C
(37)
Probability that the neutron is scattered toward C,
Prob )( own own ;; dnucleialldnucleiall upSCup =
=
2N
1
Sa iii
C
(38)
The interference among the different paths is reflected in the peaks of interference shown in Fig. 7.12.
29
Counts
Fig. 7.12
CASE-2B.2 (interchange of spin)
Initial state
Final state
S
C
1 2 3 N
1 2 3 i N
Fig. 7.13
Amplitude probability that the neutron is scattered towards
C via the nucleus-i:
iSC
s via nucleu
dnucleiallupdownunucleus i
own ;;
downnucleusother theall
p
=
= iC Sb i (39)
The difference between (35) and (39) is that in the latter case, the interaction has left a signature indicating which nucleus participated in the interaction. In other words, it can be distinguished from the event in which the neutron is scattered by, for example, the nucleus
k. (We can go and verify which nucleus made the flip without affecting
the information detected by the counter.)
Amplitude probability that the neutron is scattered towards
C via the nucleus-k:
kSkC
s via nucleu
dnucleiallupdownunucleus
own ;;
downnucleusother theall
p
=
= kC Sb k (40)
30
Each case, (39) and (40), leaves the crystal in a different configuration. This means that the final states represented in (39) and (40) are different. Since paths with different final states do not
interfere, we have to calculate first the individual probabilities iC
Sb i 2 for i = 1, 2, 3,… , N; and then add them up
Probability that the neutron is scattered towards C: Prob)( own ; p ; dnucleiallSunucleitheofoneC updown =
=
N
1
2
Sb i
iiC (41)
Since all the phases are gone, no peaks of constructive interference will be displayed; rather a smooth distribution will be detected by the counter (as shown in Fig. 7.14.)
Counts
Fig. 7.14
When cases 1 and 2 combine, the result will be a diagram like the one depicted in figure 7.15.
Neutrons diffracted by a
crystal of atoms whose nuclei
have spin 0
Nuclei of
spin 1/2
Counts Counts Neutrons diffracted by a
crystal of atoms whose nuclei
have spin 1/2
Fig. 7.15
1 Richard Feynman, “The Feynman Lecture on Physics,” Vol-III, Addison-Wesley,